Health information to help you understand symptoms. It isn't a diagnosis.
Check symptoms

Research deep dives

What drives long COVID: the leading explanations, study by study

Last reviewed November 15, 2026. OwieClear editorial team. Waiting for clinician review. Not medical advice. Talk with your doctor about your own care.

Long COVID almost certainly isn't one disease. It is a label for people left ill after a COVID infection, and researchers now think several different processes are involved, sometimes in the same person and sometimes not.

Below are the main proposed drivers with what the evidence actually shows for each, including the places where a popular idea has run into trouble. Where a finding has failed to replicate, we say so.

Related pages: Long COVID · Chronic fatigue syndrome (ME/CFS)

The bottom line

What we know
  • Long COVID is a real, physical illness with measurable abnormalities in blood and muscle tissue.
  • Post-exertional malaise, the worsening after effort, is a core feature and is not deconditioning.
  • No single mechanism explains every case.
What is still open
  • Whether leftover virus in tissues is driving symptoms or is just debris.
  • Whether microclots are real and important, or an artifact of the method. Recent work points against them.
  • How much of long COVID is autoimmune.

The studies, in order

  1. 1. Viral leftovers (Several research groups)
  2. 2. Old viruses waking (Gold 2021; Peluso 2022 and others)
  3. 3. Microclots (Pretorius and Kell from 2021; Fraser and colleagues, 2026)
  4. 4. The immune system (Many groups)
  5. 5. Nerves and signaling (Wong, Levy, Thaiss and colleagues, 2023)
  6. 6. Muscle and energy (Appelman, Charlton and colleagues, 2024)
What the evidence labels mean
GuidelineAn official recommendation from an expert group after weighing all the evidence.
Review of many studiesCombines the results of many studies (a systematic review or meta-analysis).
Randomized trialPeople were randomly given a treatment or a comparison, the best way to test cause and effect.
Large observational studyFollows many people over time. Shows strong links, but not always cause and effect.
Small or early studyA small study, case series, or first results. Needs to be confirmed.
Lab or animal studyDone in cells or animals. May not apply to people.
Study 1 of 6

Pieces of the virus stay in tissues after the swab goes negative

Several research groupsSummarized in a 2024 Cell reviewLarge observational study

What they did

Researchers looked for SARS-CoV-2 RNA and proteins in blood, gut tissue and other organs months after infection.

What they found

  • Viral RNA and spike protein have been found in gut tissue, lymph nodes and blood long after recovery.
  • The idea is that this leftover material keeps the immune system switched on.
  • In muscle biopsies from people with long COVID, viral protein was found, but at the same level as in people who had recovered fully, which suggests it is leftover debris rather than the cause of symptoms there.

What it means

This is the most testable hypothesis, because it predicts that an antiviral should help. That prediction has been tested, with disappointing results so far.

What it can't tell us

  • Finding viral material doesn't prove it's causing symptoms.
  • Studies differ on how often it is found and where.

Read it yourself

Next: Old viruses waking

Study 2 of 6

Dormant viruses, especially EBV, reactivating

Gold 2021; Peluso 2022 and othersPathogens and other journalsLarge observational study

What they did

Researchers measured signs of Epstein-Barr virus reactivation in people with and without long COVID.

What they found

  • In one study, about 67% of people with long COVID had markers of EBV reactivation, against 10% of controls.
  • A larger, more careful study also found recent EBV reactivation associated with long COVID.
  • That same study found long COVID in people with no sign of prior or recent EBV infection at all.

What it means

The picture is that COVID can loosen the immune system's grip on viruses it was already holding down. That may contribute in some people.

What it can't tell us

  • EBV reactivation is not required for long COVID, so it cannot be the whole explanation.
  • Nearly everyone carries EBV, and reactivation markers turn up in healthy people, which makes this hard to interpret.
  • The earliest study was small and used self-reported symptoms.

Read it yourself

Next: Microclots

Study 3 of 6

Microclots: a popular theory that is not holding up

Pretorius and Kell from 2021; Fraser and colleagues, 2026South Africa and the UKLarge observational study

What they did

One group reported tiny clot-like deposits in the plasma of people with long COVID that resist being broken down, and proposed they block small vessels and starve tissue of oxygen. Other groups set out to check.

What they found

  • The original work reported large numbers of these deposits, and a later study using flow cytometry reported a roughly 20-fold increase against healthy controls.
  • A 2024 Cell review noted the findings remained to be confirmed in independent groups, and that no study had yet linked clotting pathways to actual tissue damage.
  • In October 2026, an Oxford group studied never-hospitalized people with specialist-confirmed long COVID and found no evidence of persistent clotting activation, no impaired clot breakdown, no abnormal fibrin structure and no excess microaggregates, and no link to breathlessness, fatigue or lung perfusion.
  • The authors concluded blood-thinning treatment is not justified outside trials for these patients.

What it means

This matters practically, because the microclot idea is used to sell anticoagulants, enzyme supplements and blood-washing procedures. The best controlled test so far did not find the thing those treatments are supposed to fix.

What it can't tell us

  • The negative study was in people who were never hospitalized; it may not settle the question for everyone.
  • The method used to detect microclots is hard to standardize between labs, which is part of why this has stayed unresolved so long.

Read it yourself

Next: The immune system

Study 4 of 6

An immune system that won't stand down, and antibodies against the self

Many groupsLarge observational study

What they did

Researchers compared immune markers in people with long COVID and people who recovered.

What they found

  • Signs of ongoing immune activation and altered T cell populations are reported repeatedly.
  • Antibodies against the body's own tissues have been found in some studies, raising the idea that COVID sets off an autoimmune process.
  • Muscle biopsies in long COVID showed immune T cells infiltrating muscle tissue, which is not normally seen.

What it means

If this is right, drugs that calm or redirect the immune system are a logical thing to test, and several are in trials.

What it can't tell us

  • Autoantibody findings have been inconsistent across studies, with some groups failing to find them.
  • Immune activation could be a response to something else rather than the primary problem.

Read it yourself

Next: Nerves and signaling

Study 5 of 6

Serotonin, the vagus nerve, and why standing up is hard

Wong, Levy, Thaiss and colleagues, 2023Cell, University of PennsylvaniaLab or animal study

What they did

Researchers compared blood metabolites in people with long COVID, people with acute COVID, and people who recovered, then tested the mechanism in mice.

What they found

  • Circulating serotonin was reduced in long COVID. Three routes contributed: less tryptophan absorbed in the gut because of inflammation, fewer platelets to store serotonin, and faster breakdown.
  • In mice, reduced serotonin weakened vagus nerve signaling, which impaired the hippocampus and memory.
  • Restoring serotonin or stimulating the vagus nerve reversed the memory problems in mice.
  • Separately, autonomic problems such as POTS, where the heart races on standing, are well documented after COVID.

What it means

This offers a concrete route from gut inflammation to brain fog, and points to testable treatments.

What it can't tell us

  • The mechanism was worked out in mice. The human part was an association in blood samples.
  • No trial has yet shown that giving tryptophan, serotonin precursors or SSRIs treats long COVID. The researchers said a proper trial is the next step.
  • Taking supplements on the strength of a mouse study is not supported, and serotonin-raising supplements can interact with medicines.

Read it yourself

Next: Muscle and energy

Study 6 of 6

The clearest evidence: what exercise does to long COVID muscle

Appelman, Charlton and colleagues, 2024Nature Communications, Amsterdam UMC and VU AmsterdamLarge observational study

What they did

25 people with long COVID and 21 people who had fully recovered rode a bike to maximum effort. Blood and muscle biopsies were taken a week before and one day after.

What they found

  • Before exercise, the long COVID group already had more easily-fatigued muscle fiber types and lower mitochondrial function, meaning muscle cells made less energy.
  • One day after exercise, muscle tissue showed severe damage, muscle fiber death, and immune cells infiltrating the tissue, far beyond normal post-exercise soreness.
  • Blood metabolites related to energy production were severely reduced.
  • The problem was not the heart. Something was stopping muscle from using the oxygen delivered to it.

What it means

This is the strongest biological evidence that post-exertional malaise is a physical process in the muscle, not deconditioning or fear of activity. It is the direct reason pacing is advised and push-through exercise programs are not.

What it can't tell us

  • It was a small study. Published commentary argued the findings could partly reflect deconditioning and that brain-level regulation may also be involved; the authors disagree.
  • What causes the mitochondrial problem in the first place is still unknown.

Read it yourself

More plain-language reading

See something wrong or unclear on this page? Suggest a correction. OwieClear is for education only and isn't medical advice.